Could 3D Printing Lead to Distributed Manufacturing?



Additive manufacturing (AM), also known as 3D printing, is a game-changer for the field of manufacturing, enabling significant savings of cost, time and materials. In traditional manufacturing, parts are manufactured in large quantities at centralized factories, then shipped out to consumers. But with the growth of AM, many wonder whether this technology will cause a shift from this centralized model to a more distributed model, in which facilities in different locations coordinate to fill manufacturing needs.

A team of researchers from Carnegie Mellon University’s Department of Engineering and Public Policy (EPP) and the University of Lisbon investigated how AM could contribute to distributed manufacturing. They examined whether AM will disrupt this central model, specifically in the context of spare parts for the aviation industry, where being able to quickly print parts instead of stockpiling them would be attractive.

"Our results suggest that 3D printing may not be as conducive to distributed manufacturing as some might hope," said Parth Vaishnav , an assistant research professor in EPP and member of the Next Manufacturing Center. Vaishnav and his colleagues also suggest in a recent paper in the journal Additive Manufacturing that it is more suitable for non-critical parts that do not need to be expensively processed after they emerge from the AM machine.

Research Engineer Parth Vaishnav in the Department of Engineering and Public Policy discusses greenhouse gas emissions from civil aviation, 3D printing parts used by airlines, and the damage done to the atmosphere by carbon dioxide.

The team also includes Erica Fuchs , a professor in EPP; Joana Mendonça, professor at the University of Lisbon; Jaime Bonnín Roca, who was a Ph.D. student in EPP when the study began, and is now an assistant professor of innovation and entrepreneurship at Eindhoven University of Technology in the Netherlands; and Ria Laureijs, a Ph.D. student in EPP.

This research is part of the Carnegie Mellon Portugal Program , an international partnership with Carnegie Mellon and the Government of Portugal where CMU and Portuguese universities, research institutions and companies collaborate on education, research and innovation.

The team points out that the discussion around AM’s potential role in changing manufacturing practice have been exaggerated in some contexts, especially if one considers its economics. In manufacturing, a company chooses whether to site its factories nearer or farther from their markets and customers, which affects shipping and stockpiling costs. While decentralized AM close to where products are needed may reduce the need to carry inventory, the very expensive machines that are sometimes needed to process parts after they emerge from 3D printing may still favor centralized manufacturing.

To figure out how a slew of influences might shape 3D printing in industry, the research team used cost and location models to estimate supply chain costs in aviation, an industry that already 3D prints some parts. The team found that conventional, centralized manufacturing would likely remain more cost-effective when compared to distributed. Decentralized manufacturing would become attractive only if production volumes increased or part processing needs decreased.

The researchers chose to focus their research on non-critical spare parts, since the Federal Aviation Administration, which ensures the safety of the U.S. air system, only requires that such parts be equivalent to existing ones and meet necessary safety requirements. A crucial assumption in the analysis is that many different types of parts could be printed on a single 3D printing machine.

The team concluded that decentralizing manufacturing would make economic sense for non-critical spare parts with combined annual production volumes in the 10,000s.

The model suggests that this logic is unlikely to change even if the 3D printers themselves become much cheaper or faster. If the 3D printers improved enough that post-processing steps could be eliminated, then decentralized manufacturing would be economical at lower volumes-but these would still be in the tens of thousands. Cheaper machines would also mean that labor will become a larger proportion of total costs. This could be mitigated by automating the machines, which in turn could affect employment.

Additive manufacturing has enormous potential for good, and the researchers encourage companies and policymakers to take it seriously. They should, however, not assume that it goes hand in hand with distributed manufacturing.